Final Results of the Sunbelt Melanoma Trial

1
Final Results of the Sunbelt Melanoma Trial

• Kelly M. McMasters MD, PhD, Merrick I. Ross MD,
  Douglas S. Reintgen MD, Michael J. Edwards MD,
  Charles R. Scoggins MD, MBA, Robert C.G. Martin
  MD, Arnold Stromberg PhD, Lee Hagendoorn MBA,
  Marshall Urist MD, James Goydos MD, R. Dirk Noyes
  MD, B. Scott Davidson MD, Stephen Ariyan MD,
  Jeffrey Sussman MD, Peter Beitsch MD, Jeffrey E.
  Gershenwald MD
• From the Department of Surgery, University of
  Louisville, on behalf of the Sunbelt Melanoma
  Trial Group

2
Disclosure

• This is an investigator-initiated clinical trial
  supported by a grant from Schering Oncology
  Biotech
• All data management performed at University of
  Louisville
• PI and several other investigators have been on
  Schering Oncology Biotech Speakers Bureau

3
Sunbelt Melanoma Trial

• Randomized prospective trial
• Eligible patients
• Age 18-70 yrs
• Cutaneous melanoma 1.0 mm thickness
• No clinical evidence of regional nodal or distant
  metastases

4
Sunbelt Melanoma Trial

• 79 institutions across U.S. and Canada
• 3619 patients registered
• Stratification
• Breslow thickness
• 1.0-2.0 mm, gt2-4 mm, gt4 mm
• Ulceration
• Final analysis approved by DSMB

5
Sunbelt Melanoma Trial
Melanoma 1.0mm thick Age 18-70
SLN Histology Negative
SLN Histology Positive
PROTOCOL A LN Dissection
RT-PCR Negative
RT-PCR Positive
gt 1 Positive LN or Extracapsular Extension
PROTOCOL B
Obs Arm 7 N908
1 Positive Node Only
Randomize
Randomize
IFN Arm 3 N99
CLND IFN Arm 6 N184
CLND Arm 5 N192
Obs Arm 4 N180
Obs Arm 1 N112
IFN Arm 2 N106
6
RT-PCR Analysis of SLN

• Tyrosinase
• Mart1
• Mage3
• Gp100
• RT-PCR with Southern Blot Detection
• ¼ of LN or 2 mm diameter piece of LN, whichever
  smallest
• Central lab, high quality control standards
• SLN Positive test defined as detection of
  tyrosinase at least 1 other marker

7
RT-PCR analysis Southern Blot
Results
Tyrosinase
Mage-3
Mart-1
gp100
b-actin
Path
PCR


Patient 1

-
Patient 2

-
Patient 3
-

Patient 4
-
-
Patient 5
Negative
M12
8
Power Analysis
9
Protocol A Clinicopathologic Factors
10
Protocol B Clinicopathologic Factors
11
Protocol A DFS (ITT)
0.9
0.8
0.7
0.6
Surviving
0.5
0.4
0.3
0.2
0.1
0.0
0
1
2
3
4
5
6
7
8
9
10
11
DFS (years)
12
Protocol A (ITT)
13
Protocol A Arms 1-3, DFS (ITT)
14
Protocol A Arms 1-3, OS (ITT)
15
Protocol B DFS (ITT)
16
Protocol B OS (ITT)
17
Protocol B Arms 4-7, DFS (ITT)
18
Protocol B Arms 4-7, OS (ITT)
19
Arms 1-7, OS (ITT)
Arms 4,5,6,7
Arms 1,2
Arm 3
20
ECOG HDI TrialsRelapse-Free Survival Benefit of
HDIHazard Ratio by Number of Nodes Positive

• Number of E1684 E1690 E1694 nodes
  positive (ITT) (eligible) (eligible)
• 0 0.53 1.46 2.07
• 1 2.29 1.00 1.44
• 2 3 1.24 1.92 1.16
• ? 4 1.18 1.15 1.47

Hazard ratios for E1684 are based on
intent-to-treat, whereas the hazard ratios for
E1690 and E1694 are based on eligible cases. P
lt0.05
Kirkwood et al.
21
EORTC 18952 PEG-IFN LD vs. HD vs. OBS
DMFS in N1 Patients

100
90
80
70
60
50
40
30
20
Overall Logrank test p0.486
10
(years)
0
0
1
2
3
4
5
6
7
O
N
Number of patients at risk
61
144
120
99
74
52
20
5
2 YRS,LD
65
141
112
88
74
41
15
5
1 YR,HD
35
68
54
41
36
20
11
2
OBS
Eggermont et al, LANCET
Oct 2005
22
EORTC 18952 PEG-IFN HD vs. LD vs. OBS
DMFS in N2 Patients

100
90
80
70
60
50
40
30
20
Overall Logrank test p0.708
10
(years)
0
0
1
2
3
4
5
6
7
O
N
Number of patients at risk
Treatment
84
138
106
75
61
43
25
2
OBS
167
271
212
139
113
80
43
12
1 YR,HD
162
270
218
161
124
87
48
10
2 YRS,LD
Eggermont et al, LANCET
Oct 2005
23
EORTC 18991 PEG-IFN vs. OBS N1 Patients
RFS HR 0.73
DMFS HR 0.75
Hazard rate over time
0.3
0.2
0.1
PEG-IFN OBS Censored
OBS PEG-IFN
OS HR 0.88
0.0
0
1
2
3
4
5
Years
Eggermont et al.
24
EORTC 18991 PEG-IFN vs. OBSN2 Patients
DMFS HR 0.94
RFS HR 0.86
0.6
Hazard rate over time
PEG-IFN OBS Censored
0.5
0.4
0.3
Hazard rate
0.2
OBS PEG-IFN
0.1
OS HR 1,01
0.0
0
1
2
3
4
5
Years
Eggermont et al.
25
Conclusions

• Protocol A
• Results do not support the use of high dose IFN
  for patients with a single positive SLN
• Patients with a single positive SLN are an
  intermediate, not high risk group
• Patients with gt1 microscopically positive LN or
  with extracapsular extension have a significantly
  worse prognosis than those with a single positive
  LN

26
Conclusions

• Protocol B
• Results do not support the use of CLND or IFN for
  node-negative patients
• RT-PCR staging of SLN was not predictive of worse
  outcome

27
Conclusions

• Caveat
• Study underpowered to detect small differences in
  DFS and OS
• Variable benefit of IFN alfa-2b in other clinical
  trials by number of positive lymph nodes

28
Acknowledgments

• Sunbelt Melanoma Trial Investigators
• Schering Oncology Biotech